Book Title: Lost Fragments Of Spitzer Manuscript
Author(s): Eli Franco
Publisher: Eli Franco

View full book text
Previous | Next

Page 2
________________ Eli Franco Lost Fragments of the Spitzer Manuscript Hara. Two of the fragments are transcribed also on p. 28", two also on p. 35," and one also on p. 64 of the Nachlass"; I point out the variants in the notes. Further, whenever available I note the variants in WATANABE's transcriptions as published by MIYASAKA. Note that I did not change SPITZER's transcriptions and that sometimes better readings appear in the notes rather than in the main text. In the discussion below I also suggest some conjectures in pointed brackets and add punctuation to facilitate the reading. Finally, I have prepared a concordance for MIYASAKA's transcriptions which I append to this paper. mine with the help of the transcriptions that some of the remaining fragments were broken into two or three pieces after SPITZER had transcribed them, presumably during or after WW II. Relying on SPITZER's transcriptions I was able to paste together the following fragments: 8 + 623; 30 + 74 + 140: 73 +64 + 76; 139b1[7] + 139a1(1): 767 + 780; 842 +847. Since SPITZER did not transcribe the entire manuscript, it is impossible to ascertain what portion of the fragments has been saved by his transcriptions. One can only hope that he transcribed the larger fragments and that those fragments that he did not transcribe were smaller and less significant. In this connection one has to mention that the great Japanese scholar Shoko WATANABE, who worked with Ernst LEUMANN in Berlin during the thirties, transcribed about 100 fragments. WATANABE handed over his transcriptions to Yosho MIYASAKA, and the latter passed them on to Junkichi IMANISHI, currently a professor at the IABS, Tokyo. A careful comparison between SPITZER's and WATANABE's transcriptions is an urgent desideratum. On the one hand, such a comparison will allow us to better determine how many of the fragments were lost during the war. On the other hand, as will be seen below, a comparison between two independent transcriptions is bound to eliminate some mistakes that occur in the one or the other." In what follows I reproduce and discuss the transcriptions of the first three pages of SPITZER's Nachlass as a token of my respect for and friendship with Professor Minoru [Page 1] Blatt 369 (1) khalv api kin vijānīs.). ...(duh)kha[m] tathāt parisistan(i) duḥkham ca duhkh(a)tatprāptam parikiştäny api duḥkhan tasmad" duḥkhadarsankt paribistadarsa The pasting is done only virtually on the images of the digitalized fragments, of course; I did not actually remove the fragments from their glass frames. The numbers of the fragments refer to the numbering of glass frames preserved in the State Library in Berlin. Fragment No. 10, which has also been transcribed by Sprzen, was broken probably during the sixties because it appears unbroken on the black and white photos used by SCHLINGLOFF when he worked on the manuscript during the early sixties, which he has kindly put at my disposal. CEMIYASAKA, ibid., p. 673. Incidentally, some unpublished pages by WATANABE remain in LHUMANN's Nachlass, cf. Catalogue of the Papers of Ernst Lumann in the Institute for the Culture and History of Indian and Tibet, University of Hamburg. Compiled by Birte Plutat. Stuttgart 1998, entries 381, 387, 503. MIYASAKA states (ibid. p. 673) that the fragments amount to 110% leaves, but given that 854 glass frames containing fragments still remain, that some of the frames contain more than one fragment (27. I believe, is the highest number of fragments in a single frame), and that some fragments (certainly more than 40) were lost, the statement cannot be correct. If it were true, it would mean that on sverige each fragment would have been broken to nine or ten pieces, which seems highly improbable. My assumption is that WATANABE transcribed some 110 fragments and that MIYASAKA's statement refers to WATANABE's transcriptions, not to the original fragments. It is probable that MIYASAKA did not see the actual manuscript, at least not before writing his paper, because he also says that the fragments are made of birch-bark, while in fact they are made of Talipat palm leaves. I would like to thank Professor KYOMA for a translation of MIYASAKA's paper. "CE. c.g., a case of aberratio oculi pointed out in 372a2 below. P.28 contains the transcription of another small fragment which contains the word pariini and thus seems to belong to folio 369, but SPITZER did not include it there. Next to the transcription of these two fragments SPITZER refers to some inspecified location "6 unten". The pages are numbered by me in the order in which they were placed in the folder One of the most conspicuous phraseological characteristics of the SPITZER Manuscript is the frequent use of thaly api. It is interesting to note that G. VON SIMSON, who investigated the language of the wires of the Sarvistivada school, pointed out that the use of thalu is usually avoided in the Sarvistivada texts. CL. G. VON SIMSON, "Stil und Schulzugehörigkeit buddhistischer Texte." in H. Bechert (od.), Zur Schulzugehörigkeit von Werken der Hinayana-Literatur. Part 1 (Symposien zur Buddhismusforschung, II,1), Gottingen 1985, pp. 76-93, in p. 83: "Auf der anderen Seite scheint bei den Sa. der Wille zur Abkehr und Entfernung von der mittelindischen Vorlage starker ausgepragt zu sein als bei den MQ. Charakteristisch hierfür ist die weitgehende Vermeidung der Partikel khal im PrMoSD ( Anh. IL 1, Nr. 33-37), die ebenso wie die Partikel bho im Pali sonst in buddhistischen Sanskrittexten abersus gewöhnlich ist. It seems therefore, that either the avoidance of thaly was not yet common in the 3 century, or that was limited to the sätra literature and was not applied to Abhidharma commentaries. Further, there is no absolute certainty that the Spitzer manuscript belongs to the Sarvativ de school. The use of khalval rather than thalu slone seems to presuppose a different Innguage of reference than Pali. "MIYASAKA, p. 676, 2761, corrects: tarha (sa)parifistäni. Note that MrYASAKA's transcription of this line begins with ... ... kham, i.e. without khalv apikin vijānif. MIYASAKA's transcription, p. 676, differs considerably; it omits duhkham ca and reads dukhal (sic.) praptam. Read:... duhtham tar praptam. "MIYASAKA, p. 676, reads tamanna. of this Vind ASAKA S Eduhubhom minna.

Loading...

Page Navigation
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18